| 1. |
- Hasegawa, Tai, et al.
(författare)
-
Edge-Based Graph Neural Networks for Cell-Graph Modeling and Prediction
- 2023
-
Ingår i: Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings. - : Springer Nature. ; , s. 265-277
-
Konferensbidrag (refereegranskat)abstract
- Identification and classification of cell-graph features using graph-neural networks (GNNs) has been shown to be useful in digital pathology. In this work, we consider the role of edge labels in cell-graph modeling, including histological modeling techniques, edge aggregation in GNN architectures, and edge label prediction. We propose EAGNN (Edge Aggregated GNN), a new GNN model that aggregates both node and edge label information to take advantage of topological information about cellular data and facilitate edge label prediction. We introduce new edge label features that improve histological modeling and prediction. We evaluate our EAGNN model for the task of detecting the presence and location of the basement membrane in oral mucosal tissue, as a proof-of-concept application.
|
|
| 2. |
- Nair, Aravind, et al.
(författare)
-
A graph neural network framework for mapping histological topology in oral mucosal tissue
- 2022
-
Ingår i: BMC Bioinformatics. - : Springer Nature. - 1471-2105. ; 23:1
-
Tidskriftsartikel (refereegranskat)abstract
- Background Histological feature representation is advantageous for computer aided diagnosis (CAD) and disease classification when using predictive techniques based on machine learning. Explicit feature representations in computer tissue models can assist explainability of machine learning predictions. Different approaches to feature representation within digital tissue images have been proposed. Cell-graphs have been demonstrated to provide precise and general constructs that can model both low- and high-level features. The basement membrane is high-level tissue architecture, and interactions across the basement membrane are involved in multiple disease processes. Thus, the basement membrane is an important histological feature to study from a cell-graph and machine learning perspective. Results We present a two stage machine learning pipeline for generating a cell-graph from a digital H &E stained tissue image. Using a combination of convolutional neural networks for visual analysis and graph neural networks exploiting node and edge labels for topological analysis, the pipeline is shown to predict both low- and high-level histological features in oral mucosal tissue with good accuracy. Conclusions Convolutional and graph neural networks are complementary technologies for learning, representing and predicting local and global histological features employing node and edge labels. Their combination is potentially widely applicable in histopathology image analysis and can enhance explainability in CAD tools for disease prediction.
|
|
| 3. |
- Tollemar, Victor, et al.
(författare)
-
Histopathological Grading of Oral Mucosal Chronic Graft-versus-Host Disease : Large Cohort Analysis
- 2020
-
Ingår i: Biology of blood and marrow transplantation. - : Elsevier. - 1083-8791 .- 1523-6536. ; 26:10, s. 1971-1979
-
Tidskriftsartikel (refereegranskat)abstract
- Graft-versus-host disease (GVHD) can manifest as acute or chronic complications in patients after hematopoietic cell transplantation (HCT). Oral chronic GVHD (cGVHD) occurs in approximately 70% of HCT recipients and includes lichenoid-like mucosal reactions, restricted mouth opening, and salivary gland dysfunction. However, the underlying histopathological presentation remains to be validated in large cohorts. We characterized the histopathological features of oral mucosal cGVHD and devised a scoring model in a large patient cohort (n = 112). Oral mucosal biopsy sections (n = 303) with and without oral cGVHD were identified from archived and current HCT recipients with additional healthy controls. Histological screening was performed on hematoxylin and eosin-stained and periodic acid-Schiff-stained sections. A points-based grading tool (0 to 19, grade 0 to IV) was established based on intraepithelial lymphocytes and band-like inflammatory infiltrate, atrophic epithelium with basal cell liquefaction degeneration, including apoptosis, as well as separation of epithelium and pseudo-rete ridges. Validation involved 62 biopsy specimens, including post-HCT (n = 47) and healthy (n = 15) specimens. Remaining biopsy specimens (n = 199) were blindly graded by 3 observers. Histological severity was correlated with clinical diagnostic and distinctive features, demonstrating a spectrum of individual patient severity, including frequent signs of subclinical GVHD in healthy mucosa. However, oral cGVHD presented with significantly higher (P < .001) scores compared with HCT controls, with moderate to high positive likelihood ratios for inflammatory infiltrate, exocytosis, and basal membrane alterations. The grade II-IV biopsy specimens demonstrated a histopathological diagnosis of active mucosal lichenoid-like cGVHD, highlighting the importance of correlating clinical presentation with the dynamic histopathological processes for improved patient stratification. In addition, this tool could be used for assessing treatments, pathological processes, and immune cellular content to provide further insight into this debilitating disease.
|
|
